Split Factor - Grounding systems 2

[ohms]

Hi every one.

I am getting confused with the current division factor Sf for system design and the split factor for buried mesh. Can anyone shed some light on it with some references.

Many thanks for your time.

 

[jghrist]

When a ground fault occurs in a station and the source of the current is remote from the station, some current flows through shield wires or line neutral conductors back to the source. Some flows from the ground grid through the earth back to the source. The current division factor Sf is the fraction of the total fault current that flows through the earth.

 

[cuky2000]

To determine the split (current division) factor (S_f), three methods are suggested:

a) Simplified graphical method or impedance calc from Annex C IEEE Std 80-2000.
_{Example: a) Select the applicable graphic determining the fault category, tower footage resistance, etc. b) Start the horizontal axe with the calculated grid resistance and select the curve for A/B (were A= # of transmission line & B=# of secondary feeders).}

S_f could be obtained on the vertical axe

b) Compute the maximum short circuit fault current to ground (I_f) and the zero sequence current(I₀) at the desired location.
_{Example :Max[ISLG, ILLG or I3LG].}

S_f = I_g/3.I₀

c) Use specialized software such as EPRI (Substation Grounding Programs), CDEGS from Sestech or similar.
_{Beware that those programs are not inexpensive, require training and good understanding of system parameters.}

 

[ohms]

Thanks for your feedbacks.

Infact, our customer wants us to take the Sf as 100% for earthing system design. That we agreed and did the calculation considering full fault current (without considering any dicersion).

However, for Earth Mesh, we consider 70% of fault current for conductor sizing considering that the current will split in at least 2 paths when entering the ground mesh.
However, we are getting hard time to convince consultant that it is in line with standards IEEE chapter 15 and also BS 7354.
They were saying that since we have consider Sf as 1 for system design, the same to be considered for mesh conductor size.

I am of the opinion that somehow we can separately see these 2 things. i.e if we take Sf as 70%, we can still take take additional 70% while sizing ground mesh.

How you ppl advise.

 

[jghrist]

These are two separate things. Who pays for the larger conductor, you or the customer? If it is the customer, then use the larger conductor. There is no downside except for cost.

 

[cuky2000]

Conductor size has minimum impact on the grid performance.
The main concern of conductor size is to withstand the heat during the design clearing time to avoid annealing or fusing the conductor. The degradation of the conductor should be considered in a corrosive environment for the duration of the facility (25 to 30 years typ.)

The IEEE Std 80 provide guide to size the conductor
See also the enclosed link: ttp://cuky2000.250free.com/Cable_Fusing.jpg

 

[ohms]

jghrist

Infact we have to provide the conductor. :-D thats the reason we dont want to oversize it without any technical reason.

From your comments, I understand that you also see both as a separate thing and we can safely consider additional 70% for grid conductor sizing as I said in my above post. right?

ohms

 

[bacon4life]

We have some substations with split factors of .1, but it seems like a a bad idea to size the mesh using that factor. Even using your example of .7*.7 = 0.49 would require current to split into at least three conductors.

I cannot find a reference to using a mesh grid split factor in IEEE std 80. Perhaps it is in IEC standards? Although it seems reasonable to assume the current does split in the mesh, it would take some calculations to show that it is never more than 70%/30%. In addition, all equipment grounding leads would still need to be full size.

You may also need to show that the joints from the grounding leads to the smaller mesh conductor will not overheat as they could have twice rated current flowing through them.

Also the owner may not explicitly consider degradation but rather rely on somewhat initially over sized conductors.

 

[cuky2000]

Bacon4life,

There is a fairly good documentation in the IEEE Std 80-2000 (check annex C). Enclosed is a sample of current division factor.

http://cuky2000.250free.com/Gnd_Example_Current_Div_Factor.jpg

Current division factor could vary in a wide range from substation to substation. For instance, in a current 138 kV metropolitan project, the utility provided values of SLG=30kA with a current division factor of 20%. Therefore the current injected into the ground for grounding calculation purposed was able to be reduced to 6 kA.

The low current division factor was possible for the presence of a UG pipe type cable feeder @ 138 kV.

 

[jghrist]

Even if most of the fault current returns through the shield wires instead of through the earth (low Sf), you probably still need to consider the entire fault current in sizing the grid conductors. This is because the current may have to flow through the grid conductors to get from the fault point to the shield wires.

The current division in the grid conductors is mentioned in IEEE Std 80-2000, Section 11.3.3.

 

[bacon4life]

IEEE Std 80 is quite clear about the split factor for determining the whether the current returns via the mesh or the shield/neutral wire. The reference jghrist provided sure leaves a lot up to engineering judgement for what split factor would be acceptable for the mesh conductors.

 

[cuky2000]

Hi Jghrist,

The ground grid conductors will carry significantly less current than the max fault current to ground and even less than the total current injected into the ground since the current will be distributed among several return paths.

However, for risers the situation will be different requiring one or more paths to be considered without counting the help of other metallic return paths such as steel structures/anchor bolts, equipment stand, transformer tank, etc.

 

[jghrist]

cuky,

It depends on where the fault is and where the shield wire is terminated. Consider the following:

   +-----------+----------+-----------+
   |           |          |           |
   |         []|        []+ Take-off  | 
   |           |          | Tower     |
   |           |          |           |
   |           |        []+ Bus Support
   +-----------+----------+-----------+
   |           |          |           |
   |           |          |           |
   |           |          |           | Grid conductor
   |           |          |           |
   |           |          |           |
   +-----------+----------+-----------+

If the fault occurs at the bus support, current can flow both directions, but most of the current returning in the shield wire will flow in the shortest most direct path in the single grid conductor between the bus support and the take-off tower.

 

[cuky2000]

Several facts need to be considered for grounding design.
• For single bus support, even if the anchor bolts are not directly connected to the rebar, their close proximity and the semi-conductive nature of concrete will provide an electrical path. So the fault current conduction in a single bus support with ground riser connected to the ground grid is split between the riser conductor and the steel structure-anchor-foundation and soild combination.
• Conductor can withstand high level of fault without melting. For example, a 4/0 AWG conductor can withstand before fusing 42,kA for 0.5 sec , 95 kA for 0.1 sec and more than 300 kA for 0.01.
• Most substation steel members are connected to the rest of the station ground electrode with usually not less than 2 copper conductors of not less than No. 2/0 AWG in such a way that should one grounding conductor be damaged, no single metal structure or equipment frame may become isolated. So the overall probability of melting failure is minimized.
• Special care is recommended selecting the type connectors to withstand the thermal stresses. For instance the maximum allowable temperatures for connectors are as follow:
*Cu-Exothermic Weld = 1083°C
*Brazed Joint = 450°C
*Mechanical Joint = 250°C